You are here:
NRAO Public Wiki
>
HPC Web
>
CasaParallelization
>
CasaParallelMinutes20110718
(2011-07-25,
JamesRobnett
)
Edit
w
iki text
Edit
Attach
P
rint version
Casa Parallelization meeting minutes
Monday July 18th Room 317, 3:30pm MDT
Attendees: Kumar, Jim, Jeff, James, Justo
Targets/tasks for June 18th - July 24th
Create a local account for Justo
James
Test
CalMS
selection in partition
James
Fill 400GB data set
Sanjay
Discussion/Agenda
Justo visiting for 2 weeks
Working with Jim on current state of affairs
Working on async flagging
MS selection appears to be slow, wrong sort order most likely
Examine async thread failure problem
Jim/James
Memory resident subtables may help
Move subtables to in memory storage rather than storage manager
Jim
Simon Radcliffe from SKA
Do we want to track A/W projection and GPU work here or is it too early
Spectral imaging
Kumar
Parallelization status
Issue with more processes than number of MSes
fixed
Input output channel mapping is serial and slow, moving to children
Complete parallel user interface frame work
Jeff/Honglin
Memory usage increases in pcont as a function of number of cores vs spws due to data slice bug/feature
Shoot for end of month to have a proto parallel script
Miriam running into a problem where some data doesn't have calibrator and setjy fails
Need to differentiate bad selection from parallell generating null selection in exception handler
Refine and distribute cluster scheduler interface
James
In progress, awaiting proto parallel script
Potentially use OODT on order of a month
Need 400+GB continuum case with complex field
Sanjay
Sanjay will fill to an MS that others can read from
Observed last Friday
Deferred Topics
Test version of auto-flagging
deferred
Few weeks from suitable for testing
Parallel gridder testing
James
deferred
MS-MFS performance vs image size
James
Deferred
Appears to be an L1-L2 cache issue on older systems (newer Nehalems have one cache line per core)
Consider other parallelization approaches
Work toward implementing streaming framework in applycal
Jim
deferred
deferred for subtable storage manager rework
Priorities based on 3.3 release plans
Parallel framework
BDF size independent filler aka streaming filler
Async I/O
Multims spectral, continuum imaging
auto-flagging
ARDG Targets
What is negative side effect of CVel in filler, non-reversible
What needs to be done to acceptably implement Auto-boxing with large number of channels
Broader parallelization of minor cycle for MS-MFS
--
JamesRobnett
- 2011-07-25
E
dit
|
A
ttach
|
P
rint version
|
H
istory
: r1
|
B
acklinks
|
V
iew wiki text
|
Edit
w
iki text
|
M
ore topic actions
Topic revision: r1 - 2011-07-25,
JamesRobnett
HPC
Log In
Register
HPC Web
Create New Topic
Index
Search
Changes
Notifications
Statistics
Preferences
Webs
ALMA
NAASC
SCIIPT
CDL
CICADA
Ccs
Littlethings
Cville
DSAA
EVLA
FASR
GB
Computing
Data
Dynamic
Electronics
Gbtpipeline
Knowledge
Mechanical
Observing
WbandVLBACal
Obsreports
Operate
PTCS
OOF
ServoImprovementsHome
ServoSiteAcceptTestProcs
Pennarray
Projects
Recreation
Scicenter
Skynet
Software
CLEO
StaffEvents
TACTool
HPC
JVLA
KPAF
Library
Main
Metrics
NGVLA
NM
Computing
Electronics
OSAA
OSX
SRDP
Software
Algorithms
CASA
Splat
System
VLBA
Copyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding NRAO Public Wiki?
Send feedback